Thursday, July 23, 2009

Importance of Employment Centers

Jarrett has an interesting post on how LA is more like Paris with their polycentric form than a more monocentric place like New York City. I've been looking all week at LEHD data, mapping out job clusters and have noticed that many places in the United States are polycentric. This is also something Richard Layman talks about a lot as well, but in a slightly different way.

For example, the Twin Cities has a number of job clusters that could be made walkable if given a push. It's quite possible that this is a better way to look at transit possibilities, rather than the traditional hub and spoke. Jared makes this point, but the proof is in how our regions are laid out and how people already commute. I don't have the maps here now, but most of the major clusters in the twin cities draw residents from around that cluster. Meaning many people live closer to where they work than we might have thought, they just don't live close to the major center of the region, but rather thier own major cluster.

This all leads up to talking about how to fill in the centers and connect those people to thier cluster. Chris Leinberger talks about Walkable Urbanism and building up centers. You can see this in DC where places have grown up around the Metro lines. In other regions, places have grown up where there are metro lines such as Atlanta, but also have grown densely but not as walkable in other places. Many of these places could be added to and reconfigured for walking.

I once thought Phoenix would be hard pressed to change its ways. But it has really good bones and a regional grid that is almost unmatched in the United States. There are also two major places outside of downtown that could be even more dense than they are today with greater access. They could already support high capacity transit, the one area north of downtown just got attached to the new light rail line.

North Central
Camelback Road

But you also have to do it right. In my travels to Denver, I noticed that the Tech Center which has the most jobs outside of downtown has fairly lousy access to the light rail line. This place will not transform as easily as it might have with the line running straight through the center of the density existing, density you can tell was created by cars.

Denver Tech Center

These pop up in other regions as well, and usually represent the best place to connect downtown with another major job center. These corridors also make for the best starter transit lines, especially if you're having to work with the cost effectiveness measure, because you're going to get the most riders from them. Houston knows this for certain, because in connecting Downtown to the Medical Center, they were able to build the highest passenger density new light rail line in the United states.

Medical Center and Rice University

In Atlanta, it's Peachtree outside of Downtown on MARTA and Buckhead just a bit further north. The point I've been trying to make is that more of these places could be created and ultimately connected together in a web with better transit. But it's much easier to demonstrate in pictures than with just words.


Looks kind of like Arlington no?

Buckhead Station in Atlanta

Which kind of looks like Bethesda

The biggest thing I think we see here is how if there is a station, the density fills in between the lines. The Phoenix example is just density for cars, not people. This all can change though, and more centers could pop up around the region to foster more walkable urban development. These centers need to be connected by transit, and if connected, will follow Jarrett's ideal:
If you want a really balanced and efficient public transit system, nothing is better than multiple high-rise centers all around the edge, with density in the middle, because that pattern yields an intense but entirely two-way pattern of demand. If balanced and efficient transit were the main goal in Los Angeles planning, you'd focus your growth energies on Westwood, Warner Center, Burbank, Glendale and perhaps new centers in the east and south, while continuing to build density but not necessarily high rise in the middle.
This way we can accommodate the complete market for housing, not just the segment that is single family, and most can have access to quality transit. We can also cut down on VMT while serving our polycentric regions with quality transit of all types.


Matt Fisher said...

I would not support BRT as a substitute for rail anywhere in the Twin Cities, including I-35W. This isn't the best I would do, in spite of all the "realities" as BRT proponents define it.

Robert said...

Great visuals, by the way.

For the Denver picture, you didn't illustrate where LRT runs in that picture, which is just to the west (left) of the 25 freeway. The tech center jobs are almost exclusively on the east side of the freeway, so it is automatically a much longer walk than it would have been had LRT been built through the action rather than on the fringes of it.

The freeway ROW is about 15 lanes at that point and the only station for the tech center is at the top of that picture, so if one works south a ways, you're looking at a really, really long walk or a transfer.

IMHO, RTD's loads would be a lot better if it built that line with better connections to the Tech Center, even though the Tech Center itself is not very walkable.

However, because of the expense that would be involved with running street-level rail right through the DTC and the locations from which the people commute to the DTC, express bus service would be ideal for getting people to the tech center carless.

Steven said...

I think a solution would be to connect the Bellevue and Dayton Light rail stations with a streetcar line running through the Tech Center.
The right of way certainly exists and the cost of a streetcar loop through the Tech Center shouldn't be that high.

Conversely, you could look at building an Automated People Mover (APM) with stops actuall in (or adjacent to) the major buildings.

Anonymous said...

As for Houston, under your argument that two major job centers were connected, I'm surprised it has such high ridership density. Sounds like two destinations connected to each other as opposed to connecting to some origin. What am I missing?

Richard Layman said...

The other key is the creation of a transit system, as opposed to just having a line or two. Arlington's stations in VA would be worthless without being connected to other parts of the region through the rest of the metropolitan transit network (fixed rail subway system).

Morgan Wick said...

"As for Houston, under your argument that two major job centers were connected, I'm surprised it has such high ridership density. Sounds like two destinations connected to each other as opposed to connecting to some origin. What am I missing?"

The origins in between that get connected to two destinations instead of just one. But I think job centers outside downtown should be served by local streetcars and good buses at best. Most of them were created by cars and sprawl, after all, and a lot of the people tying transit to development would rather revitalize downtown first and look at areas beyond it later.

(Though in that I'm probably being influenced by Jon Von Kerszek and even he proposes a polycentric transit system for LA centered around Downtown, Hollywood, and the Westwood/Century City area...)

jarbury said...

Isn't the key to limit free parking provision in these development nodes though? Surely one of the main reasons that PT modeshare for jobs in the CBD is much higher than suburban employment centre is because parking is expensive in the CBD and generally inexpensive or free elsewhere.

So the system can certainly work, but requires that a price is put on parking.

Steve said...

The Denver tech center looks a little like South San Francisco -- dense for a car-oriented office park and near transit, but nonetheless a car-oriented office park.

MB94128 said...

If you want a survey of polycentric development take a look at Garreau's (Wash. Post reporter) "Edge City" (1991).

Edge City @
Edge City @

Granted, he is looking at nodes created by freeways. But this book identifies nodal points coast-to-coast. One could use his book as a starting point for mapping where there has been transit infill and where there could be new development or extensions.